OpenAI is rolling out its latest advanced LLM, Sol, for wide public access. Sol is considered at least on par with Anthropic’s Fable, a model whose capabilities (or ownership) impressed the White House enough that it was briefly banned from public view.
So how did these models get the OK for release? Short answer: No one is completely safe.
“Honestly, I don’t have visibility into the exact processes, so yeah, I don’t feel like I have enough information to say whether they’re adequate or not,” Mina Narayanan, a senior research analyst at Georgetown’s Center for Security and Emerging Technology, told TechCrunch. “Anthropic said they were in talks with the government and that they were developing a classifier to detect jailbreak attempts, and they’ve implemented defensive gap strategies to prevent future jailbreaks, but exactly what that dialogue looked like between the government and Anthropic and OpenAI is unclear.”
Dean W. Ball, a former Trump policy adviser who now works for OpenAI, wrote that “no one knows what the requirements are to get a license” in his newsletter last month.
Andy Konwinski, a computer scientist who co-founded Databricks, Perplexity and the Laude Institute, said he’s never spoken to anyone who understands the process, including employees at Frontier Labs. “It’s an existential problem,” he tells TechCrunch. “Security or not, it’s about who has the power to make decisions – who are the gatekeepers and decide on permissions?”
Eighteen months into the Trump administration, there is still little clarity about how to move forward, despite—or, some critics argue, because of—the industry figures that set policy. Last month, after weeks of infighting, an executive order was published outlining a roadmap for evaluating border models, but the details have yet to be fleshed out, beyond what doesn’t exist. “There will not be an FDA for AI,” Sriram Krishnan, a former Andreesen Horowitz partner who served as a senior adviser on AI at the White House until last month, told the Financial Times.
In particular, there is still no consensus on what types of models require government oversight, or which agency(s) should conduct these evaluations. So far, the Commerce Department’s Center for AI Standards and Innovation appears to be taking the lead, but the executive order directs six cabinet agencies to determine a final process by early August. What has emerged in the meantime is ad hoc at best.
OpenAI CEO Sam Altman said on CNBC that the process involved talks with officials such as Commerce Secretary Howard Lutnick, Treasury Secretary Scott Bessent and US National Cyber Director Sean Cairncross, but it is not clear who the experts who tested the models were or how they did it. OpenAI declined to share details of the government’s process with TechCrunch, but pointed to the results of several external evaluations by organizations such as UK AISI, SecureBio and Irregular in the latest model’s security map.
As with Anthropic’s Fable rollout, OpenAI showed the model to the government and select users ahead of a wider release, but we don’t know who all those users were or how they were selected. In a blog post in late June, the company said that “we don’t believe this kind of public access process should become the long-term standard,” and said it would work with the government to develop another way forward.
The background to those conversations, however, includes Altman reportedly offering as much as 5% to OpenAI’s equity for the administration’s so-called “Trump Accounts,” and OpenAI president Greg Brockman’s role as the largest publicly known donor to Trump’s midterm political operation. It is difficult for outside observers to separate these activities from the government’s apparently lighter touch approach to regulating Sol.
Amthropic’s Fable, on the other hand, was briefly pulled from wider access when the US government banned its use by foreign nationals, partly due to real concerns about users jailbreaking the model to access hacking capabilities, and partly due to personality clashes between Anthropic and the Trump administration. The threat of an export ban may also have made OpenAI more cooperative with the government’s (unknown) requests.
From an industry perspective, a hands-off approach to regulation may be nice, but one that depends on personal connections to administration officials creates uncertainty and perverse incentives.
Konwinski told TechCrunch that he worries true experts in this technology — “security researchers, alignment researchers, interpretive scientists, but also data people and people from across the board” — aren’t playing enough of a role in the model release process.
Konwinski argues that an “open common area” is the best way to balance security and innovation. He points to models such as the FDA, NIH or the national laboratories, which convene researchers, government officials and private companies to reach consensus on safety issues.
Some of that comes down to the incentives of capitalism that have motivated AI researchers for more than a decade and played out in the courtroom during Elon Musk’s lawsuit challenging OpenAI’s corporate structure. Ball points out that the nature of the AI business requires companies to recoup a large portion of their training costs soon after their models are released and are further ahead of the competition.
“While their intentions are good, there are very clear legal obligations and fiduciary responsibilities built right into the operating procedures,” Konwinski points out.
Ball argued in his submission that the way forward will depend on third-party audit organizations, licensed by the government, that will evaluate border labs’ approach to security. Konwinski is also positive about new institutional formats such as focused research organizations that could help more disinterested experts from academia and the non-profit world access and evaluate frontier models.
For now, the secrecy surrounding the development of artificial intelligence is not going away, but it will also create political challenges for an industry that Americans increasingly view with skepticism. “There’s not a sense that responsible people are driving these changes,” University of Wisconsin-Madison computer science professor Remzi Arpaci-Dusseau said last week at the Open Frontier conference.
At the same event, David Siegel, the computer scientist who founded Two Sigma, one of the most successful quantitative hedge funds, asked attendees to “imagine a situation that I think would be very bad, [where] a small number of companies control the technology; the government, in their secretive laboratories, assesses whether or not the technology is fit for use; and the general public and the scientific community don’t really have access to any of it.”
It seems we don’t have to imagine it.
When you buy through links in our articles, we may earn a small commission. This does not affect our editorial independence.